Application monitoring with
Custom metrics
Powered by one of the world's most high-performance metric agent StatsBee, our custom metrics uncover new layers of insights into application performance and user experiences. With no data sampling, global percentiles built-in and high cardinality metrics and the ability to use our power SDKs Serverless, getting started will be one of the easiest and best choices your company can make today!.
Why BeeInstant's Custom metrics?
High
cardinality
BeeInstant supports high cardinality metrics so engineers can track application performance and user experience in more detail.
Top grade
accuracy
While most monitoring tools sample your data, BeeInstant cares about each and every observation to bring you the most completed and undiluted view.
Global
percentiles
Global percentiles measure the actual customer experience at the service level. BeeInstant is designed from day one to support these most useful statistics.
Publish Custom Metrics
BeeInstant's SDKs
You can use BeeInstant's SDKs to publish metrics directly from Java or Golang codes. Tags or dimensions are supported out-of-the-box. No agent required to use the SDKs. If your application is serverless, using SDKs is the best option for publishing custom metrics.
Open-Source Tools
Using StatsD or Prometheus to publish custom metrics, switching to BeeInstant is as simple as running one command line to install StatsBee. No change required in your codebase. No data sampling. Global percentiles built-in. High cardinality metrics by default. Access to the extra layer of insights.
An Example
Using StatsBee and PyStatsD to monitor loadsong api of Karaoke service written in Flask.
import time
import statsd
from flask import Flask
from flask import jsonify
app = Flask(__name__)
c = statsd.StatsClient('localhost', 8125)
@app.route("/loadsong")
def load_song():
start = time.time()
#
# Lookup song in database
# Other processing tasks...
#
c.timing("Timing,env=%s,service=%,api=%s,genre=%s,country=%s" %
("Prod", "Karaoke", "LoadSong", get_genre(song), get_country()),
(time.time() - start) * 1000)
return jsonify({ ... })
